NodeODM GPU start error

ODM
1

When I run docker run -p 3000:3000 --gpus all -d opendronemap/nodeodm:gpu I receive the following error:

6d95bc7758ace54cc2ef2b5829fd70065f8406490c16912323e351a7ef47317d docker: Error response from daemon: failed to create shim: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: nvidia-container-cli: initialization error: load library failed: libnvidia-ml.so.1: cannot open shared object file: no such file or directory: unknown.

I followed the instructions to install nvidia-docker2, and the nvidia-container-toolkit, restarted the docker service, but I’m still having issues. Any help would be lovely.

1 Like
2

Is your user part of the docker group?

What is the output of:

id $USER
3

uid=1000(odmadm) gid=1000(odmadm) groups=1000(odmadm),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),116(lxd),1001(docker)

1 Like
4

Hmm…

Can you try adding your user to the video group and re-login/reboot to try again?

Looks like NVIDIA sometimes recommends this (I have to use this myself for Intel under X):

1 Like
5

Ok, now we’re getting a different error. Almost like it doesn’t have drivers…

docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].

I’m going to reinstall drivers and see if this is resolved just like that.

1 Like
6

After installing the cuda drivers, and rebooting the machine, I am still receiving the above error.

I actually had an issue with docker!

I installed docker during installation with Ubuntu. Even though I had restarted the service multiple times, I had to reinstall it altogether. Now it’s working with the test, and I’m going to start working on testing with processing.

1 Like
7

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.